# Content policy enforcement

Automatically analyze, flag, and mask content that violates safety policies in agent conversations.

Foldspace can automatically analyze and flag user-generated content across conversations, inputs, and output streams.

When content violates or risks violating a policy, the system attaches a policy flag and masks the relevant content in the conversation to prevent unsafe or sensitive material from being displayed.

:::note[Enterprise Feature]
Content policy enforcement is available for enterprise customers. To enable or configure moderation and context-sensitive policy handling, contact [support@foldspace.ai](mailto:support@foldspace.ai).
:::

## Supported Policy Categories

| Policy Name | Description |
| :--- | :--- |
| Dangerous Content | Content that facilitates, promotes, or enables access to harmful goods, services, or activities. |
| Harassment | Content that is malicious, intimidating, bullying, or abusive towards others. |
| Sexually Explicit | Content that is sexually explicit in nature. |
| Hate Speech | Content that is generally accepted as being hate speech. |
| Medical Information | Content that promotes, facilitates, or enables access to harmful medical advice or guidance. |
| Violence & Gore | Content that includes gratuitous or realistic descriptions of violence and/or gore. |
| Obscenity & Profanity | Content that contains vulgar, profane, or offensive language. |

## How It Works

Foldspace scans conversational and generated content in real time. If a message matches one or more of the policies above:

1. The system **flags** the violation for moderation and observability.
2. The affected text is **masked** in the conversation to maintain a safe, compliant user experience.

This ensures AI-driven interactions remain aligned with safety standards and organizational policies.